Developing Data Analysis Skills
Low grades in English Composition class tend to make educators believe students are not ready to take the ACT test. Some wonder if students should take English Composition or if they should take a lower, preparatory course first. The data you will examine lists variables of students who took both English Composition and the ACT test.
Step One: Describing the Data
- Open the resource “English GPA Variables and Output.” This contains the codebook that lists the variables used in a particular study regarding low grades in English Composition class. How many variables are there?
- Examining these variables, what are two possible explanations for students who might not be ready to take an English Composition class that would NOT be related to the ACT score?
- The first step in the data analysis process is to run and frequencies. Take a look at the frequencies first. Frequencies indicate the number and percentage of people that fall into a specific category. Using the frequency data provided, write a description of what you see in variables V1 through V4.
- Examine V5, the semester grade in English Composition. What can you say about the distribution of the grades? Did the class seem to be easy or to be hard? To answer this, use the data and the grading scale provided by the school. What are the number of students who scored in the A range, the B range, the C range, and the D range?
- Examine V6, the overall grade for the first semester. Use the same grading scale as in the previous question to create a grade distribution and describe what you see.
Step Two: Developing a Cut Score
- Now that you have described the data, begin to address the problem of a “cut score.” The first step is to make sure the ACT is an appropriate test to use for your purpose. To do this, you will want to see if there is a relationship between the students’ score on the ACT test and their grades in English composition. Write a hypothesis you might make regarding what you expect to find about the relationship between ACT test scores and the English Composition grades. When you write a hypothesis, you actually write two. You will also write a NULL hypothesis which indicates there is no relationship between the ACT test scores and English Composition grades.
- Write a hypothesis and a null hypothesis regarding the differences in GPA between gender in English Composition and the difference in genders in scores on the ACT test. Base your hypothesis on the data from the code book.
- Examine the table for descriptors V3 – V6. There are three common descriptive statistics in each table that researchers call the “central tendency.” These are scores that describe the exact center or middle of the distribution around which the other scores are grouped or clustered. The three measures of central tendency are the mean, the median, and the mode. Define mean, median, and mode and give a citation for where you get your information from.
- You will also want to examine how scores are spread out from the mean. These are measures of dispersion or variability. Three common measures of dispersion are the range, the variance, and the . Write a definition for each of these terms and give a citation for where you get your information from.
- Using the data from the descriptive tables V3 – V6, write a description of what you see and especially about performance in English Composition class.
- Look at the output and find the table that lists the correlation between ACT English score and grade in English composition. The Greek letter that looks like the small letter ‘r’ is the correlation. Correlations range from -1 to 1. The closer a correlation is to -1 or 1, the stronger it is. In addition the p value gives us significance. In social sciences, p values that are less than .05 (p < .05) are considered to be statistically significant. This means that the results of a study did not happen by chance. In this case, it would mean that the relationship between the two variables is significant. There is only a 5% chance that this relationship does not exist. How would you describe the relationship between the two variables and what does it tell you about whether the ACT test scores are related to performance in English composition class? (Be careful: a correlation does not cause something to happen. A correlation simply means that a relationship exists).
- Look at the output table labeled “T-test: Grade in English Composition Grouped by ACT English Score Group.” Describe what you see. Is there support for using 18 as the cut score to determine whether can sign up for English Composition or whether they should take a different course first? ( Use the terms you’ve learned previously in this exercise—do NOT make this harder than it is by overthinking this question.)